Multi-step Image Enhancement for Incinerator Flame Recognition

In order to address the issue of low segmentation accuracy in the weak flame region of waste incineration flame images and the potential loss of texture details at the flame edge, this study proposes an algorithm for segmenting incinerator flame images based on multi-step image enhancement. The algorithm consists of several steps. Firstly, a single-scale Retinex algorithm is employed to enhance the details of the noise-reduced image. Subsequently, an adaptive gamma correction method based on the inverse color transform is proposed to enhance the contrast between the image foreground and background, in conjunction with the inverse color transform algorithm. Finally, a 2D-Otsu algorithm is utilized to segment the image and extract the target flame region. Experimental results demonstrate that the proposed incinerator flame image segmentation algorithm based on multi-step image enhancement achieves superior segmentation performance, effectively preserving more detailed information in the weak flame region.